Summary: This research program investigates immune responses induced by viral and plasmid vectors. Results are relevant to vaccine potency and mechanisms of action. In addition, immune responses can alter the safety and efficacy of gene therapy using viral or plasmid vectors, especially in the context of readministration, either by blocking therapy or by causing immunopathology. Better understanding of such immune responses can help select predictors of clinical success as well as of adverse events, and thus contribute to improved regulatory decision-making. Our work focuses on the mouse influenza system, and especially on heterosubtypic immunity, that is immunity induced by influenza A virus of one subtype and providing cross-protection against challenge with flu A of different subtypes. Vaccine development and prevention of pandemics would be aided by a more complete understanding of the broad cross-protection against widely divergent viral strains that is observed in animals. Main projects: a) Role of IgA: The role of local, mucosal immune responses in protection against influenza, and in particular in heterosubtypic immunity, is incompletely understood. We have performed mucosal immunizations and heterosubtypic challenges in knockout mice lacking IgA, to analyze the role of this specialized mucosal antibody. Results showed these mice were capable of responses providing cross-protection of both the upper and lower respiratory tract. We are now analyzing the role of T cells in these mice by in vivo depletion. b) Cross-protection in Ig-/- mice: Many proposed vaccines (HIV, hepatitis, flu) would rely on T cell immunity when antibodies are ineffective or too narrow in viral strain specificity. We have studied the protective efficacy of T cells acting in the absence of antibody. Using mice lacking all antibodies due to disruption of Ig H and L chain genes, we have shown that immunization with live virus conferred protection against heterosubtypic challenge. In vivo depletion demonstrated that both CD4+ and CD8+ T cells were required during the challenge period for optimal protection. We are now extending these studies to include DNA vaccination of Ig-/- mice and analysis of CTL induced. c) Cross-protection in mice by double negative T cells: We have studied knockout mice lacking gamma-delta T cells along with wild type mice, and shown them both capable of heterosubtyic immunity protective against challenge. Unexpectedly, T cell depletion studies gave evidence for a role of gamma-delta cells in protective immunity. When wild type mice and gamma-delta knockouts were depleted of CD4+ and CD8+ T cells during the challenge period, the wild type but not the knockout mice had remaining heterosubtypic immunity. This suggests that gamma-delta cells play a role in protective immunity, and we are further studying this possibility. d) Immune responses induced by plasmid DNA: DNA vaccines encoding conserved influenza virus antigens reduce challenge virus replication and the resulting morbidity and mortality. We had shown previously that vaccination with DNA encoding influenza nucleoprotein and matrix (NP+M) antigens protected animals, and that either CD4+ or CD8+ T cells could function without the other subset to protect the animal. We have shown that 8 months after NP+M DNA vaccination, mice are still protected against lethal challenge. We are currently preparing new constructs with the goal of increased expression of the inserted genes and thus more potent induction of immunity. d) Potential pandemic subtypes: One advantage of DNA vaccination with conserved components is the broad range of viral variants against which it protects. We have completed a study of vaccination with DNA encoding the conserved antigens NP and M of influenza virus A/PR/8 (H1N1), followed by challenge with H5N1 viruses from the 1997 outbreak in Hong Kong. H5N1 viruses of low, moderate, and high virulence were used. NP+M DNA vaccination reduced lung virus titers of low and moderate virulence strains, and protected against lethal challenge with the moderate virulence strain. Thus, in the absence of an antigenically-matched HA-based vaccine, this approach may be a useful first line of defense against a rapidly spreading pandemic virus.